Cancer Screening Tools

How scientists are exploiting biological markers to improve the diagnosis of cancer
05 November 2013

Interview with 

Nick Coleman, Professor of Molecular Pathology, University of Cambridge.


Improvements in monitoring tumours can help us improve cancer treatments, but what about diagnosis?  Biological markers might offer us a way to detect cancer in a sample more quickly than scanning thousands of cells to look for changes that might tell us that they're bad.  To find out more, Kat Arney was joined by Nick Coleman, who is Professor of Molecular Pathology at the University of Cambridge...  

Kat -   What does a good screening test for cancer look like?  

Nick -   Well, it needs to be accurate, really, so that if a patient has a cancer the test needs to be positive, and if a subject doesn't have cancer then the test needs to be negative.  That balance between avoiding false positives, avoiding false negatives is the key to an effective test.  Obviously, it also needs to be acceptable to the patient.  You don't want to be doing something that the patient wouldn't want.  It needs to be affordable.  There are practical issues as well, but I would say accuracy is the key.  

Kat -   And it sounds such a simple thing although I know it's not - it needs to be accurate and only spot cancers and not spot non-cancers.  But what are the problems with some of the current tests we have?  What are some examples of the current screening tests that don't really answer those questions?  

Nick -   So, bowel cancer screening is one example, where at the moment, people are being asked to provide a stool sample which is then tested for the presence or absence of blood.  And that test is better than nothing, but it's far from perfect because there are patients with the cancer who do not have blood in their stool and there are patients without cancer who do.  So, we end up with fairly limited accuracy.  The requirement for lots of patients who don't have bowel cancer is to undergo some pretty invasive second stage investigation to see whether or not they generally have a cancer or not.  So, there's scope for improvement in several of the cancer screening tests that we currently have.  

Kat -   Here in the UK, we currently have national screening programmes for breast, cervical and bowel cancer.  Could we screen for all cancers and why don't we?  

Nick -   It would be nice to be able to do that, but I think we have to be realistic about the strengths and weaknesses and what currently can be done.  One example is prostate cancer where it would be marvellous to be able to detect the cancers earlier and treat the cancers that need to be treated.  But the test that is currently available is a blood-based test for a protein that's made by prostate cells.  But again, it lacks the ability to detect cancer in all patients and in particular, it is prone to false positive results so that patients without cancer in their prostate have a positive result or patients who have a cancer, are identified, but those cancers are not destined to do any harm to the patient.  This is a problem of so-called 'over diagnosis'.  That's a real issue in prostate cancer, and it's currently estimated that something like 50 men have to undergo cancer treatment to save one life from prostate cancer.  So, There's a real opportunity and need for new screening approaches in prostate cancer and others too.  

Kat -   So, tell me a bit about your work.  What are you trying to do to address these needs?  

Nick -   Well, we work in cancer of the cervix or screening for prevention of cancer of the cervix, where again, this is a test which has value of course, but also has opportunities for improvement because an individual cervical smear test - if a woman has a pre-cancer in the cervix - an individual smear test is going to be positive in only about 50 to 70% of cases.  The reason for that is that currently, it's a human activity which involves a subjective assessment.  So, decision making by individuals, trying to look for very subtle differences between a pre-cancer cell in the cervix and a normal one.  So, what we've been trying to do is identify biological differences between a cancer cell and a normal cell in the cervix, so that we can stain pre-cancer cells a different colour from their normal counterparts.  And by being able to do that, it makes it much easier for these pre-cancer cells to be seen.  It means you can do it more quickly, more accurately, more cost-effectively.  There are multiple practical benefits in being able to do that.  And so, the technique that we've developed is currently undergoing trials.  We're very encouraged by the results of them so far and we're hopeful that there will be a contribution at least from this test to cervical screening.  


Add a comment